The Good, The Bad and The Ugly

[lamford]

[hv=pc=n&n=saj92hq9432dk2caq&d=s&v=e&b=3&a=3d3s]133|200[/hv]

Teams. National event.

 

All three possible calls were selected on this board at Brighton yesterday. Do you pass, double or bid 3NT? We can assume that the auction is likely to be the same in the other room to this point.

[Cyberyeti]

Preempting style in first seat green ?

 

Not passing unless partner turns up with 7 small and out on a regular basis at these colours, but happy with either of the other two, prob X at these colours.

[Jinksy]

I don't see any reason to prefer 3N over X unless partner is a very sound preemptor, and it's hard to imagine pass being the best % action unless his preempts here will be total junk.

[Lovera]

https://en.wikipedia.org/wiki/The_Good,_the_Bad_and_the_Ugly

[Trinidad]

Dbl

 

This is one reason why you preempt.

 

Rik

[mgoetze]

I played this hand opposite a solid citizen who I assumed would always have at least 5 points in diamonds, hence I bid 3NT. Playing opposite myself, it would have been a more difficult decision.

[lamford]

I played this hand opposite a solid citizen who I assumed would always have at least 5 points in diamonds, hence I bid 3NT. Playing opposite myself, it would have been a more difficult decision.

I did a simulation with the following criteria, as I had to start somewhere. Partner has 3-8 with 7 diamonds and at least 3 points in diamonds. The overcaller has KQxxxx in spades and 11-15 HCP. I shall do a simulation with partner having five points in diamonds shortly but will wait for some more comments before giving the results. Out of interest, I asked a few international players at Brighton. Brian Senior doubled, Artur Malinowski and Alex Hydes bid 3NT, and Tom Townsend passed.

[BillPatch]

I don't understand the attraction of alternatives to the penalty double at this vulnerability at IMPs. Certainly if they sit it's the surest plus. Eve if game is on for our side down 2 vulnerable will often beat a NV game. Advancer will not often find a successful rescue in ♣, which is not probable. Possibly another case of the old observation in Punch. "It has been observed that in Britain that there are only two or three authorities on bridge. It is surprising how often I draw one of them for a partner."

[Lovera]

Usually when you must decide in competitive bidding how 1♥-(1♠)-3♥-(3♠)-? if to continue bidding or double there is a rule called "of 7" and referred about number of your trumps (here 4) at which to add level getting :if it makes 7 or more tells double. [in the italian version of whikipedia is told that :"the film is than based on 3 roles arlecchino(=arlecquin or ♦) picaro(=picche or ♠) and cattivo (=bad or double)"].(Lovera)

[Vampyr]

In retrospect I should perhaps have opened 1♦. This may backfire as partner will expect me even more to have a defensive trick to add to his four; but then again it may get us to 3NT.

[BillPatch]

Unfortunately, I do not subscribe to the bulletin of the Brighton tournament. Therefore, give us the hands, so we can see how much a risk your dear English experts took by bidding the NV game rather than the Vulnerable penalty. Not my cup of tea.

[lamford]

Unfortunately, I do not subscribe to the bulletin of the Brighton tournament. Therefore, give us the hands, so we can see how much a risk your dear English experts took by bidding the NV game rather than the Vulnerable penalty. Not my cup of tea.

https://app.pianola.net/Results/Session57909/Travellers/19 is the relevant link from the excellent Brighton Hub which uses Pianola to show all the results.

 

As you can see there were plenty of scores of -730 and +400 or +430 for NS. Of course one swallow does not make a summer and my simulation gave an expected score of +489 for double, +227 for 3NT and +131 for the supine pass. I lost the maximum of 16 IMPs when doubling the cold 3S against the 3NT bid on the same auction in the other room!

 

And in response to mgoetze, when I do the simulation again so that partner has at least five points in diamonds, it does not make much (relative) difference, the relevant figures being +599 for double, +377 for 3NT and +194 for pass. 3NT almost always makes, but is usually 3 or 9 in. Interestingly if partner has the ace of diamonds, 3SX only made once in 100 hands. Bridge Analyser gives a figure of +5.88 IMPs for double compared with 3NT. There were many 800s in the file I generated.

[PhilKing]

https://app.pianola.net/Results/Session57909/Travellers/19 is the relevant link from the excellent Brighton Hub which uses Pianola to show all the results.

 

As you can see there were plenty of scores of -730 and +400 or +430 for NS. Of course one swallow does not make a summer and my simulation gave an expected score of +489 for double, +227 for 3NT and +131 for the supine pass. I lost the maximum of 16 IMPs when doubling the cold 3S against the 3NT bid on the same auction in the other room!

 

And in response to mgoetze, when I do the simulation again so that partner has at least five points in diamonds, it does not make much (relative) difference, the relevant figures being +599 for double, +377 for 3NT and +194 for pass. 3NT almost always makes, but is usually 3 or 9 in. Interestingly if partner has the ace of diamonds, 3SX only made once in 100 hands. Bridge Analyser gives a figure of +5.88 IMPs for double compare with 3NT.

 

It was noble not to wait for Patch to dig a bigger hole. FWIW, Ireland's finest also passed, as did my team mate, but it clearly boils down to preempting style. The passes were somewhat predicated on trying to avoid a huge loss, since both teams were a near lock for the final (this was the last match of qualifying).

 

Personally, I hate pass. But that's mainly because I prefer strong preempts.

[lamford]

It was noble not to wait for Patch to dig a bigger hole. FWIW, Ireland's finest also passed, as did my team mate, but it clearly boils down to preempting style. The passes were somewhat predicated on both teams trying to avoid a huge loss, since both were a near lock for the final (this was the last match of qualifying).

I note that one player, A Muridae, who wore both belt and braces, allowed 4S to play undoubled! It is interesting what you say about pre-empting style, as I would expect to make 3NT far less often opposite myself, but I would still expect to beat 3S over 90% of the time. I estimated, when talking to Mike Bell just after the round, around 40% for 2 off, 50% for 1 off and 10% of making, and this is about right, even opposite a pile of rubbish. Give partner 6-7 diamonds and 2-5 points, and 3S now makes 14% of the time, but the expectancy is still +297 for double and now -58 for 3NT and +78 for pass. Interestingly the average gain from doubling compared with 3NT goes UP as partner's hand gets worse and you now gain 7.64 IMPs per board.

 

So, pre-emptive style does not matter very much, and you should be more inclined to double if you are aggressive. That is why you are aggressive! And some of the hands that make 3Sx might well have overcalled 4S as well.

[gwnn]

Why not compute some IMP expectations (3 of them I suppose -- p/X, p/3N, X/3N) instead of expected aggregate score?

 

auto x for me.

[lamford]

Why not compute some IMP expectations (3 of them I suppose -- p/X, p/3N, X/3N) instead of expected aggregate score?

 

auto x for me.

I did compute the IMP expectation between x/3NT for two scenarios, partner has 5-8 with seven diamonds and partner has 2-5 with six or seven diamonds. The gains were given as 5.88 IMPs and 7.74 IMPs respectively. The gains over pass would be greater in the first case and less in the second case, but I did not run those.

 

I agree the auto x, and was astonished that anyone would consider anything else.

[Vampyr]

I did compute the IMP expectation between x/3NT for two scenarios, partner has 5-8 with seven diamonds and partner has 2-5 with six or seven diamonds. The gains were given as 5.88 and 7.74 respectively. The gains over pass would be greater in the first case and less in the second case.

 

I agree the auto x, and was astonished that anyone would consider anything else.

 

Well. I don't think it is "auto" to double with four tricks when you need five, especially when you know that partner is considerably less likely than normal to produce a defensive trick. But perhaps I am not looking at this hand objectively, since I was seriously annoyed at the time.

[lamford]

Well. I don't think it is "auto" to double with four tricks when you need five, especially when you know that partner is considerably less likely than normal to produce a defensive trick. But perhaps I am not looking at this hand objectively, since I was seriously annoyed at the time.

As no doubt were the partners of the other 19 doublers who conceded -730.

[ahydra]

I doubled on this hand and, of course, conceded -730. Partner said I should have opted for 3NT as doubling partscores at teams is risky business (and even if we do beat it, it might only go for 200) while 3NT was almost certainly cold - a valid point, but I felt rather unlucky having hit one of the very small number of layouts where 3SX makes. (Declarer was void in diamonds, had a 5-4 club fit with the CK onside for him, and dummy showed up with two spades and two hearts opposite declarer's AKx so the heart loser was ruffed away immediately)

 

After the event partner and I identified 3 calls I'd made which were particularly suboptimal (to put it politely). He rated this the worst of the three so it was a (pleasant?) surprise to see people and even a simulation agreeing with my X. I'll e-mail partner this link and see if he changes his mind :)

 

ahydra

[Finch]

We can assume that the auction is likely to be the same in the other room to this point.

 

Really? Why?

I was only aware of the auction at four tables (two matches) before I read some of the replies to this thread, and at those tables there were 3 different opening bids selected and the modal one was not 3D.

Pre-empting style is hugely relevant. Opposite my regular partner I would pass 3S and not consider it even close; I'd simply be pleasantly surprised if 3S trickled one off.

 

Full disclosure (you can see it on the traveller if you look): I didn't concede 730 on this hand; I conceded 790. Before you laugh too much at me, I might observe that (i) defeating 4Sx is not entirely trivial at the table (although obviously it is easy on a forum) and (ii) following on from PhilKing's comment: this was the last board we played and we were also a virtual lock for the final at that point, I possibly didn't give the play as much attention as I should have done.

[gnasher]

There really should be a carryover.

[PhilKing]

There really should be a carryover.

 

Even with no carry-over, it's still important for your NGS rating. ;)

[lamford]

Really? Why?

I was only aware of the auction at four tables (two matches) before I read some of the replies to this thread, and at those tables there were 3 different opening bids selected and the modal one was not 3D.

Pre-empting style is hugely relevant. Opposite my regular partner I would pass 3S and not consider it even close; I'd simply be pleasantly surprised if 3S trickled one off.

We have to make certain assumptions or the reader cannot realistically make a decision. The auction was the same in both rooms up to this point in our match. I am sure that some will have opened partner's hand 4D or 1D, but those will not present the same problem.

 

If I give partner a forced hand such as x xx QJxxxx Jxxx which is about as bad as he can have, then doubling still rates to be a winner, although 3NT fades into obscurity of course. Opposite this junk, with the parameters I set, for better or for worse, the distribution is 0 4 85 11 0 starting with six tricks, so usually just one off. Of course, declarer play and defence is DD. Doubling compared with pass now gains only 2.42 IMPs. Did you really let through -790? Then you should clearly have been the second pair to defend 4S undoubled! Perhaps you were keeping your energy for the final, for which my congratulations.

[akwoo]

1) I'm surprised by that simulation result (w partner having x xx QJxxxx Jxxx), but not so surprised as to not believe it. I'm not too worried about the double dummy aspect there, because it really doesn't make too much of a difference. (Surely declarer will play you for the HQ and most spades if you double!)

 

2) The hand I'm worried about partner having isn't that one; it's x Kxx JTxxxxx xx (yes that's a favorable preempt in my book, at least against good opponents), where my lead of the diamond king blows the setting trick. (It might be instructive to do DD analysis specifying the opening lead of the DK.)

[rhm]

https://app.pianola.net/Results/Session57909/Travellers/19 is the relevant link from the excellent Brighton Hub which uses Pianola to show all the results.

 

As you can see there were plenty of scores of -730 and +400 or +430 for NS. Of course one swallow does not make a summer and my simulation gave an expected score of +489 for double, +227 for 3NT and +131 for the supine pass. I lost the maximum of 16 IMPs when doubling the cold 3S against the 3NT bid on the same auction in the other room!

 

And in response to mgoetze, when I do the simulation again so that partner has at least five points in diamonds, it does not make much (relative) difference, the relevant figures being +599 for double, +377 for 3NT and +194 for pass. 3NT almost always makes, but is usually 3 or 9 in. Interestingly if partner has the ace of diamonds, 3SX only made once in 100 hands. Bridge Analyser gives a figure of +5.88 IMPs for double compared with 3NT. There were many 800s in the file I generated.

I am a fan of simulations, but in this case your simulation result will simply reflect your assumptions.

Do you assume diamonds to run?

Passing can hardly be right then.

My personal preferences are that a three level minor preempt opposite an unpassed partner should show a non-solid suit, but which should run opposite a doubleton top honor.

But this is nowadays hardly standard at these colors.

What about DBL?

Trouble is it depends a lot how short overcaller will be in diamonds and how long his spades will be.

Whatever you assume one thing is clear:

The shorter the diamonds the more attractive an overcall will look red versus white and the less attractive DBL becomes.

DBL is probably the long term winning action, the occasional double game swing against you notwithstanding.

The actual layout looks to me very unfortunate.

 

Rainer Herrmann